Re: BUG #16109: Postgres planning time is high across version - 10.6 vs 10.10

Andrei Lepikhov <a.lepikhov@postgrespro.ru>

From: Andrey Lepikhov <a.lepikhov@postgrespro.ru>
To: Mukesh Chhatani <chhatani.mukesh@gmail.com>, Andres Freund <andres@anarazel.de>
Cc: pgsql-bugs@lists.postgresql.org
Date: 2019-11-13T06:58:28Z
Lists: pgsql-bugs, pgsql-hackers

13.11.2019 00:01, Mukesh Chhatani пишет:
> Thanks for getting back to me so quickly
> 
> Queries are same and executed in 2 different environments. There are no 
> long running queries on any of the environments since they are idle 
> right away for my testing.
> 
> I can try full vacuum if that is recommended since I tried vacuum 
> analyze on the whole database in both environments.
> 
> Datases have the same content and size is also same.
> 
> Sorry but I am never seen this before , if the sizes vary or if the 
> content varies I have seen execution time being slow and not the 
> planning time.
> 
> I am working on a dataset which I will share for further investigation 
> and analysis.
Interesting. I will be waiting for your data set.

-- 
Andrey Lepikhov
Postgres Professional
https://postgrespro.com
The Russian Postgres Company



Commits

  1. Include information on buffer usage during planning phase, in EXPLAIN output, take two.

  2. Include information on buffer usage during planning phase, in EXPLAIN output.

  3. Expose BufferUsageAccumDiff().